Amazon Unveils Graviton4: A 96-Core ARM CPU with 536.7 GBps Memory Bandwidth

https://ift.tt/au85dUF

Nowadays many cloud service providers design their own silicon, but Amazon Web Services (AWS) started to do this ahead of its rivals and by now its Annapurna Labs develops processors that can well compete with those from AMD and Intel. This week AWS introduced its Graviton4, a 96-core ARM-based system-on-chip (SoC) that promises to challenge renowned CPU designers and offer unprecedented performance to AWS clients.

"By focusing our chip designs on real workloads that matter to customers, we are able to deliver the most advanced cloud infrastructure to them," said David Brown, vice president of Compute and Networking at AWS. "Graviton4 marks the fourth generation we have delivered in just five years, and is the most powerful and energy efficient chip we have ever built for a broad range of workloads."

The AWS Graviton4 processor packs 96 cores that offer on average 30% higher compute performance compared to Graviton3 and is 40% faster in database applications as well as 45% faster in Java applications, according to Amazon. Given that Amazon did not reveal many details about its Graviton4, it is hard to attribute performance increases to any particular characteristics of the CPU.

Yet, NextPlatform believes that the processor uses Arm Neoverse V2 cores, which are more capable than V1 cores used in previous-generation AWS processors when it comes to instruction per clock (IPC). Furthermore, the new CPU is expected to be fabricated using one of TSMC's N4 process technologies (4nm-class), which offers a higher clock-speed potential than TSMC's N5 nodes.

"AWS Graviton4 instances are the fastest EC2 instances we have ever tested, and they are delivering outstanding performance across our most competitive and latency sensitive workloads," said Roman Visintine, lead cloud engineer at Epic. "We look forward to using Graviton4 to improve player experience and expand what is possible within Fortnite.”

In addition, the new processor features a revamped memory subsystem with a 536.7 GB/s peak bandwidth, which is 75% higher compared to the previous-generation AWS CPU. Higher memory bandwidth improves performance of CPUs in memory intensive applications, such as databases.

Meanwhile, such a major memory bandwidth improvement indicates that the new processor employs a memory subsystem with a higher number of channels compared to Graviton3, though AWS has not formally confirmed this.

Graviton4 will be featured in memory-optimized Amazon EC2 R8g instances, which is particularly useful to boost performance in high-end databases and analytics. Furthermore, these R8g instances provide up to three times more vCPUs and memory than Graviton 3-based R7g instances, enabling higher throughput for data processing, better scalability, faster results, and reduced costs. To ensure security of AWS EC2 instances, Amazon equipped all high-speed physical hardware interfaces of Graviton4 CPUs.

It should be noted that Amazon offers up to 64 vCPU cores with its R7g instances (i.e., one full Graviton3 CPU). Therefore, to offer 192 vCPU cores, Amazon will need to either install two 96-core CPUs into one server (which marks one of the world's first Armv9-based 2-way server design), enable three-way SMT on a 96-core CPU, or implement a very low-latency high-bandwidth interconnection between servers running 96-core cPUs.

Graviton4 R8g is currently in preview, these instances will be available widely in the coming months.

Sources: AWS, NextPlatform

from AnandTech https://ift.tt/5pGJbKw
via IFTTT

Best PC Power Supplies: Black Friday 2023

https://ift.tt/n1SpfKL

Now that you've picked out your CPU, it's time to start picking out the rest of your system components. And perhaps the most humble but overlooked of these components is the power supply unit (PSU). Available in a wide range of sizes and power capacities, there are a number of great PSUs out there, but choosing between them can be a challenge. So today we're bringing you our annual PC power supply guide, to help you sort figure out what the best options are, be it a low-wattage unit for a small form factor PC, or a hulking kilowatt unit for the most powerful PC.

from AnandTech https://ift.tt/THIX1OG
via IFTTT

AMD's 96-Core Ryzen Threadripper Pro 7995WX Hits 6.0 GHz on All Cores with LN2

https://ift.tt/ZUAreBG

AMD's top-of-the-line Ryzen Threadripper Pro 7995WX is a unique workstation-grade processor that not only offers unbeatable performance in programs that can take advantage of all of its 96-core might, but it also supports overclockability. As it turns out overclocking potential of the CPU is quite huge as it can hit 4.80 GHz on all cores with air cooling and 6.0 GHz on all cores with liquid nitrogen.

AMD's Ryzen Threadripper Pro 7995WX is a quite formidable processor that packs 96 Zen 4 cores clocked at 2.50 GHz by default (and which can push one core all the way to 5.10 GHz with decent cooling), an eight-channel DDR5 memory subsystem, and 128 usable PCIe 5.0 lanes. But AMD decided to show that the processor can do much better with advanced and extreme cooling. To do so, the company teamed up with Bill 'Sampson' Alverson, an accomplished overclocker from the U.S., who took an Asus Pro WS TRX50-Sage WiFi motherboard, four G.Skill DDR5-6400 CL32 memory (which he set to work at 3200 MT/s CL32 mode), and an Asus Thor 1.5 kW power supply.

At first, Sampson tried to overclock AMD's Ryzen Threadripper Pro 7995WX using a giant IceGiant ProSiphon Elite air cooler with four huge fans. He managed to hit 4,791 MHz in Cinebench R23 and 4,875 MHz in Cinebench R15. Then, he took a custom liquid cooling system and pushed the 96-core CPU to 4,966 MHz in Cinebench R23; 5,000 MHz in Cinebench R15; and a whopping 5,265 MHz in GPUPI 1B for CPU.

But the most interesting part is of course usage of liquid nitrogen cooling as −196°C (− 320 °F) certainly removes most of thermal constraints. The results are quite impressive: Sampson has managed to push AMD's Ryzen Threadripper Pro 7995WX all the way to 5,550 MHz with LN2 cooling, which is over 584 MHz higher compared to his result with custom liquid cooling. The CPU scored 186,756 points in Cinebench R23, up from 100,000 points for the CPU at default clocks. The processor consumed 980W at such an extreme frequency, which is up dramatically from its 350W TDP. Meanwhile Elmor and Der8auer have managed to overclock his Ryzen Threadripper Pro 7995WX to 6.0 GHz wih LN2 and scored 201,501 points in Cinebench R23.

Meanwhile, to somewhat simplify overclocking botg overlockers had to reduce the number of memory modules to two and lowered its data transfer rate to 3,000 MT/s, which somewhat reduced the value of such overclocking (after all, nobody uses this CPU with only two memory modules). Meanwhile, considering the fact that AMD wanted to achieve the maximum clock possible for its 96-core processor, playing with memory configuration is a fair thing to do.

from AnandTech https://ift.tt/mf2Xev9
via IFTTT

AMD Ryzen 8000 CPUs could be revealed at CES 2024 – but not the chips you think

https://ift.tt/ZLAvlVd

Gigabyte just told us that AMD’s next-gen APUs for the desktop will be out later in January.

A press release from the motherboard maker was published over the weekend for its new beta BIOS release for AM5 (the chipset for current-gen Ryzen 7000 processors).

That release mentions the incoming desktop APUs specifically, saying: “The forthcoming AM5 next gen APU will be launched at the end of January 2024.”

If true, with CES 2024 happening at the start of January, we’d expect AMD to hold an initial reveal at this event.

These processors are purportedly Ryzen 8000G models, as previous chatter from the grapevine has indicated. Note the ‘G’ which means they are APUs with relatively powerful graphics (for an on-board solution) intended to be used in desktops so a discrete GPU isn’t needed (saving money).

In theory – as reported by HKEPC previously – the top-end Ryzen 8000G model will extend to eight cores (16-threads) and will offer integrated RDNA 3 graphics with 12 compute units.

Purportedly there’ll be four chips in the line-up, all based on Zen 4 architecture (but also incorporating more compact Zen 4c cores in a hybrid design).

Speaking of saving money on GPUs, another option, of course, is to keep an eye out on the bargains incoming for Black Friday.

---

Analysis: Pepping up the graphics

While Ryzen 8000 has been previously mentioned a lot in the context of Zen 5 processors – the next generation to succeed Ryzen 7000 – as noted, these ‘G’ suffixed APUs are rather different in their focus on providing an all-in-one desktop solution (dispensing with the need for a separate GPU).

The Ryzen 8000G models have been given considerably more pep (in theory) thanks to more powerful integrated graphics, with the shift to RDNA 3 being a big step up.

Now, while we say that, don’t think that these chips will work for serious gaming needs – but they’ll be fine for more casual gamers, delivering palatable enough 1080p performance in general. Demanding gaming will still require one of the models at the top of the tree in our best processors roundup.

The successor to the current-gen Ryzen 7000 processors, based on Zen 5, are going to turn up a bit later next year – we’re not sure when, but probably mid-year, most likely – and they may be called Ryzen 9000 models to differentiate them. That’s the latest rumor anyway.

Via PC Gamer

You might also like

• What is a graphics card? Here's everything you need to know
• These are all the best cheap graphics card deals
• Need a 1440p graphics card? These are the top options

from TechRadar: computing components news https://ift.tt/PlZRx3k
via IFTTT

Report shows Intel’s 14th-gen CPUs are great overclockers – and suggests which 14900K variant you should buy

https://ift.tt/YpznbqD

Intel’s new Raptor Lake Refresh CPUs have been under scrutiny in a new report which evaluated a host of silicon from the 14th-gen range.

This comes from Igor’s Lab – a German site which has previously conducted similar tests – and involved some 600 of Intel’s processors, both 14th-gen and 13th-gen for comparison. All these chips were split up among five intrepid overclockers for testing (one of those being Igor).

Given the vast number of CPUs being covered, the testing employed was necessarily limited, with the processors placed in an Asus Maximus Z790 Hero motherboard, and Silicon Prediction (SP) values recorded (indicating the quality of the chip essentially).

Furthermore, the voltage drawn was measured when at the maximum rated Turbo clock speed (with a lower voltage meaning the chip is better quality – as there’s more headroom to crank up the voltage for a bigger overclock).

There’s a mountain of results to wade through, as you can imagine, so we’ll pick out some brief highlights.

The overall conclusion is that Raptor Lake Refresh processors are better binned (of a higher silicon quality) compared to their predecessors, Raptor Lake chips, but then you’d expect that really (given generational process and manufacturing refinements).

For example, looking at maximum voltages, the 14900K needed as low as 1398mV in four cases to hit full Turbo, and the lowest the 13900KS managed was 1433mV (25 of the 14900K CPUS were below that value, in fact). This shows that with specs otherwise identical, the 14900K is a superior piece of silicon overall (though there will be variances, and it’s important to note the 13900KS sample size is much smaller here, which could have an impact).

As mentioned, with more headroom to allow voltage to be pushed in general, Raptor Lake Refresh is giving overclockers more scope for juicing up clock speeds that little bit more.

Another interesting point is that the report compares the standard 14900K against the 14900KF version (which is the same, but lacks integrated graphics). Was there any marked difference in the two spins on the flagship?

The 14900KF chips had a spread of SP values ranging from 90 to 104, whereas the 14900K was between 92 and 112, indicating a higher quality for the latter.

(Image credit: Future / John Loeffler)

---

Analysis: No alarms and no surprises…

Looking at those last SP scores, the 14900K looks a better proposition, but we have to remember that even a sample of 600 processors hardly tells a definitive story, whether that’s pertaining to individual head-to-head CPU performances, or the overall picture.

That said, as noted it’s not surprising to see Raptor Lake Refresh is better binned than the previous-gen Raptor Lake chips, and indeed that the new CPUs overclock better. If you recall, it was even the case that long before the 14th-gen processors emerged there were leaks indicating that Raptor Lake Refresh was going to be a solid step forward on the overclocking front.

So really, there are no big surprises here, but the result of that 14900K versus 14900KF battle is interesting to see.

Still, any Raptor Lake Refresh buying decision will be based on how much of an upgrade it’ll be from what you’re currently running, and the performance improvements for the 14900K aren’t impressive gen-on-gen as we pointed out in our full review.

If you’re intending to overclock, though, there is definitely a bit more mileage in this new generation of Intel silicon, and the report certainly makes this clear (just make sure you get yourself a top-notch CPU cooler).

Via VideoCardz

You might also like

• Processor upgrade: how to choose the right CPU
• Best gaming CPU: 5 top processors reviewed and rated
• Here are the best cheap processor deals right now

from TechRadar: computing components news https://ift.tt/vnzCqSs
via IFTTT

Is AMD close to revealing new Ryzen 9000 CPUs? We doubt it

https://ift.tt/08oTE5W

We’ve caught a glimpse of AMD’s Ryzen 9000 range, or at least a mention of it, purportedly in an Alienware advert.

Tom’s Hardware brought our attention to the leaked image which shows an Alienware desktop PC with text indicating that it has an ‘AMD Ryzen 9000 Series’ chip inside.

Alleged AMD Ryzen 9000 Listing Hints at Incoming Zen 5 Family https://t.co/7olhau8CUF pic.twitter.com/yiE6EWp8ErNovember 9, 2023

See more

Predictably, some folks have been pretty quick to jump to the conclusion that this means next-gen Zen 5 chips called Ryzen 9000 are coming, and maybe sooner than expected. And that AMD will skip over the Ryzen 8000 name (sort of – we’ll come back to that later).

However, we must be very skeptical about the image for a number of reasons, not the least of which is that it comes from the Chiphell forums (not the most reliable source in our experience).

So next up, let’s mull over the other reasons to approach this one with caution, and decide exactly how much weight we can give this apparent leak.

---

Analysis: Fake or mistake? Or could it be real?

The first thought that may occur to you is that this is a fake – that was certainly the case for us. As mentioned, Chiphell isn’t our favorite trusted source by any means, and there’s something about the ad that feels a bit off.

Alternatively, it could be a genuine piece of Alienware marketing which someone has blighted with a typo, so it shouldn’t read ‘Ryzen 9000’ at all. It wouldn’t be the first time something like this has happened.

For a moment, let’s put those two obvious concerns aside and assume that it’s authentic. Isn’t it simply too early for next-gen Ryzen 9000 CPUs to be even floating around in early work for ads? And wouldn’t it be Ryzen 8000, anyway?

On the second point, well, AMD could be reserving Ryzen 8000 for APUs and skipping it for desktops, moving straight to Ryzen 9000 with the latter. This happened with Ryzen 6000, which was APUs, where desktop CPUs went from the 5000 Series to 7000, so it’s not without precedent. Some might say it’s even quite likely.

As to whether it’s just too early to be seeing anything on next-gen Zen 5, yes, there’s some merit to that line of thinking we’d say. From the balance of rumors we’ve heard in the latter part of this year, Zen 5 is likely to debut around the middle of 2024. However, some chatter has indicated the CPUs could come earlier next year. And if we run with that idea, this leak could herald an AMD announcement of Ryzen 9000 CPUs at CES 2024, with a launch of these challengers for our best processors roundup happening a couple of months later.

That said, we haven’t witnessed much in the way of leakage around Zen 5 chips – what we have seen has been pretty lightweight spillage – so that leads us to believe it’s a real reach to assume they’re near being revealed. All in all, we’re not convinced by this leaked pic, and we’d advise staying on the skeptical side for sure – but it may well be correct that Zen 5 desktop turns out to be Ryzen 9000.

You might also like

• What is a graphics card? Here's everything you need to know
• These are all the best cheap graphics card deals
• Need a 1440p graphics card? These are the top options

from TechRadar: computing components news https://ift.tt/VR5v9Yl
via IFTTT

Micron Introduces 128 GB DDR5-8000 RDIMMs with Monolithic 32 Gb Dice

https://ift.tt/htVeAoa

The path to high-capacity RDIMMs for servers has primarily been through 3D stacking (3DS) of DRAM dice using Through-Silicon Vias (TSVs). However, this has presented significant challenges in packaging (driving up the cost), and has also not been efficient in terms of energy consumption. The demand for large memory capacity RDIMMs is being primarily driven by the sudden emergence of large-language models (LLMs) for generative AI and increasing CPU core counts. Both of these require significant amount of DRAM to keep pace with performance requirements. Keeping these in mind, Micron is introducing 128 GB DDR5 RDIMMs capable of operating at up to 8000 MT/s today, with mass-production slated for 2024.

Micron has recently started fabricating 32 Gb monolithic DDR5 dice using its proven and mature 1β technology. The new dice have a 45%+ increase in bit density, and are capable of reaching up to 8000 MT/s while also operating with much more aggressive timing latencies compared to the standard JEDEC specs. The company is claiming that it improves energy efficiency by as much as 24% compared to the competition's 3DS TSV offerings, and the faster operation can also help in faster AI training times. Avoiding 3DS TSV allows Micron to optimize the data input buffers and critical I/O circuits better, while also reducing the pin capacitance on the data lines. These contribute to the reduced power and improved speeds.

Micron has been doubling its monolithic die density every 3 years or so, thanks to advancements in CMOS process as well as improvements in array efficiency. The company sees a clear path to 48 Gb and 64 Gb monolithic dice in the future with continued technological progress. Micron is also claiming that its 1β node has reached mass production ahead of the competition, and that it has had the fastest yield maturity in the company's history. Dual-die packages and tall form-factor (TFF) modules using 1β DRAM are expected to enable 1TB modules in the near future.

Along with the announcement of the 128 GB RDIMMs using 1β technology, the company also laid out its roadmap for upcoming products. HDM and GDDR7 are expected to dominate bandwidth-hungry applications, while RDIMMs, MCRDIMMs, and CXL solutions are in the pipeline for systems requiring massive capacity. LPDDR5X, and LPCAMM2 solutions going up to 192 GB are expected to make an appearance in power-sensitive systems as early as 2026.

from AnandTech https://ift.tt/f2vpJPS
via IFTTT

Best CPUs for Gaming Holiday 2023

https://ift.tt/SF8kBy7

As we prepare to move into the holiday period, the tinsel, glitz, and fanfare of festivities are on their way, but the CPU market looks pretty solid as we round out the end of the year. Intel recently launched their latest 14th Gen Core family to the market, with the Core i9-14900K spearheading things with 6.0 GHz out of the box. While impressive on paper, there's AMD's Ryzen 7000 series to contend with, with competitiveness at the top end of the market scorching right now, and that's not a pun. Stock levels look good on processors at popular retailers such as Amazon and Newegg, and many motherboards/DDR5 memory are floating around to satisfy demand. The CPU market, on the whole, looks strong as we move into the final stages of 2023.

Looking at a wider and broader overview of the current CPU market, the top end is dominated by the new Intel Core i9-14900K (8P+16E/32T) and AMD's Ryzen 9 7950X (16C/32T). An honorable mention does need to be made regarding the Core i9-13900K (8P+16E/32T), which, although it hails from Intel's previous 13th Gen family, is nearly identical to the new 14900K bar a bump to core frequencies. All three of these processors are similarly matched in compute performance, whereas the only 'real' gain separating them is in AVX-512/AVX2 workloads, as AMD's Ryzen 7000 does support these instruction sets. Other than that, it's very competitive at the top.

from AnandTech https://ift.tt/xXrKipy
via IFTTT

Buyers of AMD Ryzen 7000 CPUs and RX 7000 GPUs get a free copy of Avatar: Frontiers of Pandora

https://ift.tt/tPqNWke

AMD is running another freebie game offer for those who are purchasing one of its current-gen processors or graphics cards.

Buyers of selected Radeon RX 7000 GPUs and Ryzen 7000 CPUs can avail themselves of a fresh bundle that provides a free copy of Avatar: Frontiers of Pandora alongside the hardware.

Wait a minute, you might be thinking – Ubisoft’s game isn’t out yet – and you’d be right there.

Avatar: Frontiers of Pandora comes out on December 7, so you won’t be able to play the game until then.

Note that only certain participating (major) retailers will carry the bundle, and when you get the GPU or CPU, you’ll need to put it in the PC where the code for the free game will be used.

The promotion ends on December 30, 2023, and you must redeem your serial code to get the game by January 27, 2024, or it’ll no longer be valid.

There’s also the small matter of Avatar-themed AMD RX 7900 XTX graphics cards with a custom shroud, and 500 of these have been made. They won’t be sold, but instead dispersed via various giveaways and promotions here and there from AMD and Ubisoft.

Naturally, it’s also worth bearing in mind that we should see some great deals on the best graphics cards out there, from AMD, Intel, and Nvidia, for Black Friday soon.

---

Analysis: Taking the baton from Starfield

The Avatar: Frontiers of Pandora bundle replaces the existing Starfield offering, and as you may realize, it’s part of the regular procession of game bundles that GPU makers throw in to give an extra incentive to buy their products.

How good will Avatar: Frontiers of Pandora be? Obviously, we don’t know yet, and won’t know until its release in a month. But it looks promising, taking cues from Far Cry and offering a two-player online co-op mode to lend some additional depth (co-op experiences are always very welcome in our books).

If you’re hankering after one of those Avatar-inspired RX 7900 XTX GPUs, then you best keep an eye out on all the usual social channels for AMD and Ubisoft for when giveaways pop up. Or you could also keep watch on eBay, where we’ll inevitably see somebody selling these limited-edition graphics cards at some point.

Via Wccftech

You might also like

• What is a graphics card? Here's everything you need to know
• These are all the best cheap graphics card deals
• Need a 1440p graphics card? These are the top options

from TechRadar: computing components news https://ift.tt/lqmFI1U
via IFTTT

AMD's EPYC 7003 "Milan" Receives Extended Lifecycle: Availability To 2026 With 6 New SKUs

https://ift.tt/4uIrcoG

AMD has sent word over this morning that the company is extending the product lifecycle for their previous-generation EYPC 7003 “Milan” family of server CPUs. The company will now be offering the Zen 3-based CPUs through at least 2026, and as part of that extension plan, has already launched a further 6 SKUs. With these changes, Milan is effectively being repositioned as a budget server platform for AMD, offering AMD’s past tech at a discount to cost-weary customers.

First launched in early 2021, AMD’s EPYC 7003 “Milan” platform is based on the company’s Zen 3 architecture, offering chips with up to 64 CPU cores, 8 channels of DDR4 memory, and PCIe 4.0 connectivity. The platform has since been superseded by AMD’s numerous Zen 4-based EPYC 9004/8004 server platforms – Genoa, Bergamo, and Siena – but for continuity and support reasons, the Milan platform is still being offered to customers. And now, it seems, it will be for even longer than AMD was first expecting.

As of today, AMD is extending the availability of the Milan platform to 2026. All the while, the company is repositioning the platform as their newest budget offering, pitching it at customers who are on the trailing-edge of server technology and need a proven platform to upgrade to, but for whatever reason aren’t ready to upgrade to the 4^th generation EPYC platforms. In practice, AMD is trying to entice Intel Xeon Scalable customers who are looking to replace their aging servers, but aren’t ready or eager to move on to 4^th gen Xeon Scalable (Sapphire Rapids).

Competitively, Milan doesn’t offer much of note with regards to performance or features compared to AMD and Intel’s contemporary platforms. But for customers already invested in it and who need the ability to repair or bring online additional matching systems over the next couple of years, it will remain in service.

Meanwhile, the company is also re-launching 6 new Milan SKUs to underscore its position as a budget platform. All of these chips were first released by AMD back in September without much fanfare, but now that AMD is extending Milan’s availability to 2026, they are using the moment to spotlight their existence.

AMD EPYC 7003 Milan New SKUs
AnandTech	Core/ Thread		Base Freq	1T Freq	L3 Cache	PCIe	Memory	TDP (W)	cTDP (W)	Price (1KU)
7663P	56	112	2000	3500	256MB	128 x 4.0	8 x DDR4-3200	240	225-280	$3,139
7643P	48	96	2300	3600	256MB			225	225-240	$2,722
7303	16	32	2400	3400	64MB			130	120-150	$604
7303P	16	32	2400	3400	64MB			130	120-150	$594
7203	8	16	2800	3400	64MB			120	120-150	$348
7203P	8	16	2800	3400	64MB			120	120-150	$338

The 6 new chips are a bit of an eclectic mix. The top two SKUs – the 56-core 7663P and 48-core 7643P – are 1P-only versions of existing Milan chips at a lower price. Meanwhile the 7303 and 7203 parts are new, with AMD offering ultra-budget server chips with just 16 and 8 cores respectively. The 7203/7203P are true entry-level chips by server standards, with just 8 CPU cores, 64MB of L3 cache, and peak clockspeed well below AMD’s only other 8 core Milan (72F3). However, all of these chips come with the full I/O and memory channels supported by the Milan/SP3 platform, allowing AMD to position them as alternatives to the more limited EPYC 8004 Siena platform.

from AnandTech https://ift.tt/u4Rkd7L
via IFTTT

Apple M3 Ultra CPU might blow away M3 Max – but we won’t find out for a long time yet

https://ift.tt/jLpfhRw

Apple’s M3 Ultra chip could be quite some way off even though the company launched a clutch of fresh silicon right out of the gate this time around.

This is according to Mark Gurman, who drops some info on the future Apple SoC in his newsletter for Bloomberg (as noticed by 9to5Mac).

As you likely saw, Apple launched the vanilla M3 alongside M3 Pro and M3 Max processors, with the latter two being a surprise (normally there’d be a gap between the release of the basic model and these turbocharged spins on the SoC).

Gurman notes the reason for this triple launch was that Apple needed to have the Pro and Max variants for use in the high-end MacBook Pros that the company unleashed at its recent ‘Scary Fast’ event.

The well-known leaker further explains that there’s only a certain amount of 3nm silicon available for Apple to use, most of which goes towards the high-demand iPhone, so the company needed lower volume products for the new M3 chips – pricey MacBooks, in other words.

Gurman then notes that the Mac Studio and Mac Pro are even slower selling products, but they would need a successor to the M2 Ultra – and this hasn’t yet gone into ‘broad testing’ we’re told.

---

Analysis: An ultra-long wait, then?

In other words, Apple is still a long way off releasing an M3 Ultra, and even though we got the M3 Pro and Max immediately, it’s likely going to be a long wait before we see an Ultra spin on the new silicon.

When the M3 Ultra arrives, if Apple sticks to its current way of doing things, the chip is likely to be very impressive, as it’ll presumably be a pair of M3 Max SoCs bolted together (as is the case with the M2 Ultra). And that current top-end offering, the M3 Max, is nicely peppy already, whereas the M3 Pro is not that much of an improvement over the M2 Pro, so Apple is favoring making the gap wider between its middling and high-end silicon, it seems.

A new Mac Studio is underway at Apple, Gurman tells us, but the next M3 products – arriving in 2024 – should be MacBook Air models (13-inch, 15-inch) and a revamped Mac mini. Notably, the leaker doesn’t say that a Mac Pro is in the works, even, so that could be even further out than any Mac Studio.

You might also like

• The 3 big ways the Apple M3 chips will change Mac gaming forever
• M3 MacBook Pro rushed out as Apple’s solution to flagging sales?
• Apple Silicon is now more important than Apple design

from TechRadar: computing components news https://ift.tt/YIpbKTJ
via IFTTT

Leaked benchmark suggests Apple’s M3 Pro CPU isn’t nearly as ‘scary fast’ as M3 Max

https://ift.tt/Po7pLmC

Apple’s new M3 Pro is not that much faster than its predecessor, if an early (leaked) benchmark is to be believed (so apply skepticism appropriately as we head into this story).

This comes from Vadim Yuryev of YouTube channel Max Tech, who tweeted (as flagged by MacRumors) to point out a spilled benchmark of the M3 Pro (the only version of the new SoC, which has 12-cores).

Breaking: The first Apple M3 Pro chip benchmarks have leaked! It’s actually slower than the M2 Max based on margin of error. M3 Pro: 15,173M2 Max: 15,242Feeling sorry for M3 Pro buyers…So embarrassing! pic.twitter.com/mdCrUnphDXNovember 5, 2023

See more

The Geekbench 6 result shows the M3 Pro – seemingly inside a new MacBook Pro 14-inch – hitting scores of 3,035 for single-core and 15,173 for multi-core.

Compare that to the M2 Pro variant with 12-cores and we see that while the M3 Pro has a good lead of 14% for single-core, when it comes to multi-core, the new SoC is just 6% faster.

Of course, that’s not a huge difference, and as a generational uplift, it’s about as small as you could get away with.

While both chips are 12-core processors as noted, the M3 Pro obviously benefits from a new architecture and more advanced manufacturing process (3nm) – but the M2 Pro benefits from having a split of 8 performance cores and 4 efficiency cores, whereas the M3 Pro has 6 of each type.

With a pair of extra performance cores, it seems that the old M2 Pro narrows the gap quite considerably – at least in certain scenarios.

---

Analysis: Don’t leap to conclusions yet

There are key points to bear in mind here. Firstly, this is just a leak, so as mentioned we should be skeptical around it. Secondly, if true, it’s still just a single benchmark, and not necessarily representative of other results we’ll see, or indeed real-world performance for the M3 Pro.

Still, this is a disappointing early hint that the M3 Pro might underwhelm at least in some respects. Particularly as with the M3 Max, Apple seems to have pushed the chip to be seriously fast (scary fast, even), to be more or less equivalent to the M2 Ultra CPU in fact (add salt again, as these are also leaked benchmarks we’ve seen for the Max).

So, there’s seemingly a pretty big gap between the M3 Pro and M3 Max, more so than with the previous generation of Apple’s silicon. And as you might imagine, some folks have been quick to jump on this as a possible tactic to get buyers of Macs to fork out for M3 Max models, rather than M3 Pro, pushing up Apple’s profits as a result.

It is still early to be drawing any such conclusions, though, and we really need to get these new laptops in ourselves and do some thorough testing before we can discern the true relationship, performance-wise, among Apple’s clutch of M3 chips.

You might also like

• The 3 big ways the Apple M3 chips will change Mac gaming forever
• M3 MacBook Pro rushed out as Apple’s solution to flagging sales?
• Apple Silicon is now more important than Apple design

from TechRadar: computing components news https://ift.tt/9FOWJks
via IFTTT

Samsung's Next-Gen 3nm and 4nm Nodes on Track for Mass Production in 2H 2024

https://ift.tt/RVt89UA

Samsung is set to start making chips using its 2^nd Generation 3nm-class (SF3) manufacturing technology as well as performance-enhanced version of its 4nm-class (SF4X) fabrication process in the second half of next year, the company told its investors this week. Both production nodes are expected to significantly improve competitive positions of the company as they will enable it to build new types of products.

"We will strengthen our technology competitiveness through the 2H [2023] mass production of the 2nd generation 3nm process as well as the 4^th generation 4nm process for HPC," a statement by the company reads. "The market is forecast to shift to growth thanks to a rebound in mobile demand and continued growth in HPC demand."

Samsung's forthcoming SF3 process technology is a major upgrade to its existing SF3E production node, which, based on the available knowledge, is only used to make small chips for cryptocurrency mining. Samsung asserts that the SF3 will offer greater design versatility by allowing for different gate-all-around (GAA) transistor's nanosheet channel widths within the same cell type. 

Although Samsung does not make direct comparisons between SF3 and SF3E, it says that SF3 offers major improvements over SF4 (4LPP, 4nm-class, low power plus), which includes a 22% increase in performance at equivalent power and complexity, or alternatively, a 34% reduction in power consumption at the same frequency and transistor count, along with a 21% reduction in logic area. In general, it is expected that SF3 will be a better option for complex designs than SF3E, though it will enter high-volume production in the second half of 2024.

In the meantime, Samsung Foundry's 4nm-class generation nodes continue to evolve. The company gearing up to launch its SF4X (4HPC, 4 nm-class high-performance computing) process technology, tailored for high-performance applications such as CPUs and GPUs used in data centers. This one is going to be SF's first process node in recent years that has been specifically designed for high-performance computing (HPC) applications.

Samsung's SF4X process technology promises to deliver a 10% improvement in performance along with a 23% reduction in power consumption. While Samsung has not disclosed the baseline for this comparison, it is likely in reference to their standard SF4 (4LPP) process. This enhancement is achieved through a redesign of the transistor's source and drain regions, following a reassessment of their performance under potentially high-stress conditions, the application of advanced transistor-level design-technology co-optimization (T-DTCO), and the implementation of a new middle-of-line (MOL) architecture.

With refined MOL architecture, SF4X is expected to achieve a silicon-validated minimum operating voltage (Vmin) for CPUs at 60mV, reduce the variability in the off-state current (IDDQ) by 10%, ensure stable high voltage (Vdd) operation above 1V without compromising performance, and enhance process margin for SRAM.

from AnandTech https://ift.tt/ZA0mBjO
via IFTTT

AMD Unveils Ryzen 7040U Series with Zen 4c: Smaller Cores, Bigger Efficiency

https://ift.tt/tQo3ZMb

In May, AMD unveiled their Ryzen 7040U series of CPUs designed for thin and light notebooks, which they codenamed 'Phoenix.' Initially, AMD announced four Zen 4-based CPUs for the ultra-portable notebook market, with the Ryzen 7 7840U (8C/16T) and Ryzen 5 7640U (6C/12T) both leading the charge with AMD's integrated Ryzen AI, which is AMD's first-generation NPU for on-chip AI and inferencing.

Fast forward a few months, and AMD is unveiling an additional two new Ryzen 7040U series processors. But there's a twist: these chips are using their smaller and highly optimized Zen 4c cores. Although AMD's more petite Zen 4c core initially debuted with the launch of their EPYC 97x4 'Bergamo' processors for native cloud deployments, server chips were never going to be the only place AMD used their compact cores. Zen 4c was already silently released in the consumer market as part of the Ryzen Z1 (non-extreme) CPU used in ASUS's ROG Ally handheld (2x Zen 4 + 4x Zen 4c), and now Zen 4c is getting a proper welcome in the consumer market with its use in a set of Ryzen laptop chips.

The first of the two new Ryzen 7040U processors with Zen 4c is the Ryzen 5 7545U, which coincidentally shares similar specifications to the existing Ryzen 7540U processor. The critical difference is that the Ryzen 5 7545U uses two full-fat Zen 4 cores with four of the smaller Zen 4c cores, which is an interesting implementation, to say the least. The second is the Ryzen 3 7440U, a direct successor to the...well, processor with the same name, the Ryzen 3 7440U, which has the exact specifications, with the only difference coming in using three Zen 4c cores, with just one Zen 4 core.

There's some nuance to pick apart with AMD's decision to use Zen 4c within their mobile platforms, as AMD highlighted at the launch of Bergamo that the 'c' in Zen 4c stands for 'Cloud,' and it's interesting to see AMD opting to integrate Zen 4c in a primarily low-end consumer segment. There are very notable differences between Zen 4 and Zen 4c, which makes things interesting, to say the least, especially when comparing the two core architectures and their intended use cases.

from AnandTech https://ift.tt/N7DMf0t
via IFTTT